Just a quick note, the decentralized swarm movement is also known as the Boids algorithm, an Artificial Life program developed by Craig Reynolds in 1986.
Great post. I would like to add that the real advantage of swarm behaviour in nature is not the behaviour itself, but the intelligence it can produce. See:
Really sharp breakdown of the distinctions here. The stigmergy comparison is spot-on, like how sequentail FPV strikes basically follow breadcrumbs from earlier hits instead of genuinely coordinating in realtime. I dunno if defensive swarms around helicopters get enough attension yet, but the APS analogy makes a ton of sense when thinking about disaggregating protection across cheaper platforms.
In the film Battle for Los Angeles, as the alien invaders in LA send their mobile HQ aloft in the face of a ground threat it is protected by 'drones' acting as weapons' platforms against the Marines, as interceptors against Marine aircraft, and also as kamikaze drones against guided projectiles lazed onto the target by the Marines on the ground. These 'drones' show the multiple roles they can tale on and appear to be networked so as to rapidly respond to the evolving threat. I remember at one point the US army was talking about shared tactical picture via imagery and hi speed burst datalinks between tanks, aircraft, and ground troops - I assume that that picture has proven harder to achieve, expensive, and in need of a number of tweaks to be effective on the current battlefield.
I imagine weight and battery life are bigger limitations than any technical difficulties. Just to act in a swarm, a drone would need at least 6 cameras or some other kind of direct signaling to maintain proper spacing. That's a drain on the battery - and hence range - and also eats into the payload. Might not be economical below a certain size, in which case a true swarm would get expensive fast.
I would examine other sensor technology than cameras and optical systems. I wonder if communications beams could double as sensors for relative positioning or could a swarm operate in an electromagnetic field that supported a single central control unit nearby to coordinate the swarm?
Possibly, I just don't see how you'd do it without radiating too much power. A centralized control unit could work, but would be a single point of failure and would also be susceptible to jamming.
For more information about the algorithm and its history I suggest:
https://www.red3d.com/cwr/boids/
Just a quick note, the decentralized swarm movement is also known as the Boids algorithm, an Artificial Life program developed by Craig Reynolds in 1986.
Great post. I would like to add that the real advantage of swarm behaviour in nature is not the behaviour itself, but the intelligence it can produce. See:
https://open.substack.com/pub/democura/p/swarm-intelligence?utm_campaign=post&utm_medium=email
Fascinating, thanks.
Really sharp breakdown of the distinctions here. The stigmergy comparison is spot-on, like how sequentail FPV strikes basically follow breadcrumbs from earlier hits instead of genuinely coordinating in realtime. I dunno if defensive swarms around helicopters get enough attension yet, but the APS analogy makes a ton of sense when thinking about disaggregating protection across cheaper platforms.
In the film Battle for Los Angeles, as the alien invaders in LA send their mobile HQ aloft in the face of a ground threat it is protected by 'drones' acting as weapons' platforms against the Marines, as interceptors against Marine aircraft, and also as kamikaze drones against guided projectiles lazed onto the target by the Marines on the ground. These 'drones' show the multiple roles they can tale on and appear to be networked so as to rapidly respond to the evolving threat. I remember at one point the US army was talking about shared tactical picture via imagery and hi speed burst datalinks between tanks, aircraft, and ground troops - I assume that that picture has proven harder to achieve, expensive, and in need of a number of tweaks to be effective on the current battlefield.
I imagine weight and battery life are bigger limitations than any technical difficulties. Just to act in a swarm, a drone would need at least 6 cameras or some other kind of direct signaling to maintain proper spacing. That's a drain on the battery - and hence range - and also eats into the payload. Might not be economical below a certain size, in which case a true swarm would get expensive fast.
I would examine other sensor technology than cameras and optical systems. I wonder if communications beams could double as sensors for relative positioning or could a swarm operate in an electromagnetic field that supported a single central control unit nearby to coordinate the swarm?
Possibly, I just don't see how you'd do it without radiating too much power. A centralized control unit could work, but would be a single point of failure and would also be susceptible to jamming.